An Exchange Theory Model of Interpersonal Communication

ABSTRACT - A new model of interpersonal communication is proposed based on social exchange theory. This conceptualization is proposed as an alternative to the prevalent "two-step" model and is offered in order to encourage new research directions.


Hubert Gatignon and Thomas S. Robertson (1986) ,"An Exchange Theory Model of Interpersonal Communication", in NA - Advances in Consumer Research Volume 13, eds. Richard J. Lutz, Provo, UT : Association for Consumer Research, Pages: 534-538.

Advances in Consumer Research Volume 13, 1986      Pages 534-538


Hubert Gatignon, The Wharton School, University of Pennsylvania

Thomas S. Robertson, The Wharton School, University of Pennsylvania


A new model of interpersonal communication is proposed based on social exchange theory. This conceptualization is proposed as an alternative to the prevalent "two-step" model and is offered in order to encourage new research directions.


The consumer behavior literature on interpersonal communication is dominated by the Katz and Lazarsfeld (1955) "two-step" model. This conceptualization represented a major contribution in dispelling the "direct effects" model, which was prevalent in the late 1940s and early 1950s and which attributed considerable unmediated power to the mass media. Katz and Lazarsfeld's research recognized the importance of horizontal communication among peer groups in mediating the effects of mass media.

Nevertheless, Katz and Lazarsfeld may also have overstated the role of social processes, especially for low involvement products where information-seeking and social legitimation may be minimal and direct media effects may be common (Krugman 1965; Ray 1973; Robertson 1976). They may also have overstated the role of social processes for innovators, who must rely to a greater extent on mass media sources of information about innovations--given the lack of information among peers (Bass 1969; Gatignon and Robertson 1985).

The two-step model is very one-sided in its structure. It suggests a dominant opinion leader in-touch with the mass media who exerts homophilous influence on a set of passive followers. The limitations of this model are well-documented (Weimann 1982; Gatignon and Robertson 1985) but central to these limitations is the one-way direction of influence. It assumes only opinion giving; however, opinion seeking and opinion sharing are also prevalent.

The model which we propose (Figure 1) has its foundations in social exchange theory (Romans 1961; Blau 1964, 1974). The central construct is reciprocity; that is, social interaction will continue only if mutually rewarding. In Blau's view: "Social limited to actions that are contingent on rewarding reactions from others and that cease when these expected reactions are not forthcoming" (p. 6). As related to interpersonal communication, influence is an exchange relationship with associated costs and benefits for both parties. The continued occurrence of personal influence is contingent upon an exchange equilibrium whereby both parties benefit approximately equally from the influence transaction.


Host personal influence for consumer products and services occurs at an informal peer group level. There may be exceptions to this, particularly for visual influence (such as fashion and automobiles) which spans diverse social environments. It might be suggested, how ever, that even visual influence is most powerful at an informal peer or reference group level, rather than among strangers. For high technology innovations consumers may have to go beyond the boundaries of their peer group in order to find adequate information and expertise. In professional and industry domains, these consumers may be more likely to seek influence in an industry-wide social domain (Czepiel 1974; Leonard-Barton 1985).

The stability of personal influence over time depends on reciprocity. An individual who supplies information to another person obligates the recipient. The recipient must, therefore, furnish benefits in return. These benefits may be tied to providing information for other product categories, since most research suggests the situational nature of influence; that is, opinion leadership varies by product category. The reciprocal benefits might also be based on voicing explicit gratitude in order to reduce the level of debt.

Figure 1 schematically represents our exchange theory model of interpersonal influence. The left side of the model takes the role of the influencer and examines the benefits and costs of providing influence; the right side takes the role of the influence; and suggests the associated benefits and costs of seeking information from personal sources. The center of the model conceptualizes a set of factors which enhance the probability of interpersonal influence.


The transmission of influence is a function of the cost/benefit analysis by the potential influencer. This analysis is rarely explicit, as in economic transactions, but is subject to the same rules of utility maximization as in economic decisions. Indeed, Blau (1974) has suggested that diminishing marginal utilities apply, in that if an influencer is constantly asked for information, the social value of being asked declines over time. Similarly, from the point of view of the influencee, initial advice is worth more than later advice (diminishing marginal utility), assuming equivalent information content.

The potential rewards from assuming the influencer role relate to decision support and justification and to social status and power. The potential costs include the social obligation incurred, the time commitments mate to information giving and the risks of providing inappropriate advice.

Benefits of Information Giving

A major incentive for assuming the influencer role is to gain support and justification for a purchase decision. In a sense, the information giver may be seeking the legitimation of friends in order to overcome cognitive dissonance. Previous research has tended to show that information giving is at its height immediately following purchase and declines with the passage of time.

A second major benefit from information giving is social status and power. The act of information giving places the influencer in a superior position. By providing information, the giver makes a claim for recognition and status. Influences, in turn, incur obligations and must demonstrate gratitude for information received or provide other services in return.

Costs of Information Giving

Much as influences incur a social cost for receiving information, influencers may also incur a social cost "for being listened to." If the influencer has been given free rein to demonstrate knowledge in one domain of consumption, it might be expected that he or she should reciprocate by receiving knowledge in another domain of consumption. Alternatively, any social relationship which is one-sided will be unstable, since the costs of information for the receiver will be too high. This will lead to resentment at constantly being in a subordinate Position and to a breakdown in communication.



There are also time commitment costs associated with information giving. These costs mag be high in a work environment, as studied by Blau, where "experts" may be accessed unduly and information giving may come to be resented. It is conceivable that this could also occur in certain high technology areas of consumer behavior. Rogers, Daley and Wu (1982), for example, found that the most likely influentials (persuaders) for personal computers had the highest levels of experience/expertise. Such individuals would seem least likely to receive purchase decision justification benefits and might be less interested in social status benefits. If so, the costs of continually providing personal computer advice could quickly exceed the rewards.

Finally, there are risks associated with the provision of information. These risks include the probability of inappropriate advice for a particular recipient and the resulting problems if recipients hold the influencer accountable. These risks may be assessed to be high for friends whose utility functions are unclear or for new technologies where long-run performance is unclear.


The reception of information is also governed by cost/benefit analysis. The potential rewards relate to the value of the information, the relief of decision anxiety, and the social definition provided. The potential costs include the risk of poor information and the assumption of a subordinate position.

Benefits of Information Seeking

The most obvious benefit of soliciting information is the value which it provides for decision-making. Information from personal sources may be of particular value if objective non-social information is lacking or is conflicting. It may also be perceived that interpersonal communication is less biased than change advocate information, such as advertising and sales personnel.

Interpersonal communication may also relieve decision anxiety. The recipient may be able to gain confidence from the prior experience of peers, or may be able to avoid extensive information seeking and processing by relying on personal influence. Blau (1974) has suggested a decision value in simply being able to "think-out-loud" with a peer.

Information seeking from friends may also be important in defining group standards for the recipient. This suggests that various consumption domains are socially defined, as in fashion and automobiles. The individual may check with relevant others in order to determine the appropriateness of certain consumption decisions. Such information seeking, or social checking, will be most prevalent when the individual has high identification with the group or a high desire for assimilation with the group.

Costs of Information Seeking

As in all information seeking, the information gained may be poor; that is, erroneous, biased or incomplete. The decision value of personal sources for many products may be overstated tue to the small sample sizes of experience which informal opinion leaders generally possess. This raises the interesting question of the relationship between opinion leadership and expertise. In consumer research on stereos, Jacoby and Hoyer (1980) found a strong positive correlation between opinion leadership and expertise. In research on home computers, Rogers,

Daley and Wu (1982) found that the opinion leaders most likely to persuade later adopters were those who were most experienced and highest on expertise. In research with a professional group (dentists), Leonard-Barton (1985) focused on national experts rather than local opinion leaders, implicitly suggesting that friendship-based opinion leadership may not correlate highly with expertise in this professional domain. The Leonard-Barton research is useful also in focusing on negative influence, whereas most research on interpersonal communication focuses on positive recommendations.

Information seeking also incurs the cost of assuming a subordinate position. Individuals may have different levels of tolerance for taking this position, but for almost all consumers it will be intolerable in the long-term unless reciprocity occurs. Such reciprocity could result from taking the opinion leadership role in other consumption domains or from more general patterns of social exchange and reciprocity, not necessarily tied to consumption.


The relative influence of interpersonal information exchange is moderated by the set of factors specified in the center of Exhibit 1. The essence of these factors is that the potential for influence is a function of the level of motivation in information search combined with the level of incentive in information giving. The degree of influence depends on the perceived attributes of the source and the perceived nature of the communication message.

Perceived Source Characteristics

Two aspects of the source explain the extent to which an individual influences others. The first deals with the credibility of the source, and the second with the attractiveness of the source. Both notions have been developed in communication theory. However, the individual nature of interpersonal communication adds new elements to the theory, beyond those which are typically pursued in the mass communication context. In general, the degree of interpersonal influence increases as the source is perceived as more attractive and as the source credibility increases.

Source attractiveness. The attractiveness of a source is determined by the individual's prestige, similarity to the receiver, and physical attractiveness. The mechanism by which attractiveness enhances interpersonal influence is the greater attention paid to the communication (Sternthal and Craig 1982).

Source credibility. The source credibility literature borrows more from cognitive theories. A highly credible source is typically more persuasive than a less credible source (Brock 1965) because the high credibility of the source inhibits counter-argumentation (Sternthal, Dholakia and Leavitt 1978). [Although Sternthal, et. al. (1978), have hypothesized an interaction of source credibility with prior opinions, a less credible source would be more persuasive if the receiver's prior opinion is positive because it would generate a greater support argumentation in the receiver's cognitive responses.] This corresponds also to the implications of a Bayesian model of information integration (Gatignon 1984). The two main determinants of credibility are the power of the source and the strength of ties between the source and the influencee.

The communication literature has concentrated on the power of the source as the major determinant of source credibility (Aaker and Myers 1982). The main type of power that has been studied is the expertise level of the source. It is clearly an essential determinant of the credibility of a source as the uncertainty associated with information is lower for a knowledgeable source than a novice. However, although the source is an expert, the information must be communicated at a technical level which can be understood by the recipient. Therefore, personal influence will be more readily accepted if the information is communicated at the technical level at which the potential recipient is knowledgeable and comfortable.

Other types of power, such as coercive power, can also moderate the degree of influence of interpersonal communication when dealing with communication between parents and children or between group members (including family) where certain levels of coerciveness can exist. This type of power might not lead to a greater credibility of the source, but can determine the extent to which the information will be used by the recipient of the information.

Further determinants of the credibility of a source can be derived form the concept of the strength of ties (Granovetter 1973, 1983). The strength of an interpersonal tie is defined as the "combination of the amount of time, the emotional intensity, the intimacy (mutual confiding), and the reciprocal sources which characterize the tie" (Granovetter 1973, p. 1261). The stronger the tie, and therefore the greater the emotional intensity and intimacy between two individuals, the greater the credibility of each of these individuals to the other. According to Granovetter, the strength of the interpersonal tie is greater between individuals who are similar and between individuals that interact frequently. Consequently, personal influence will be more readily accepted from sources who are perceived to be similar to the recipient. Also, personal influence will be more readily accepted from sources who interact frequently with the recipient. These similarity and frequency of interaction concepts are, in fact, indicators of a more general level of social integration which provides an individual with access to information available within the social system. More specifically, the credibility of the source depends on the past experience that a recipient has with the advice and information given by the source. Therefore, personal influence will be more readily accepted from sources who have low variance in their previous recommendations. In other words, potential recipients will extrapolate from their previous experience and the accuracy of prior recommendations by the source (Barone and Byrne 1984).

Communication Characteristics

A message contains information which is being given to the recipient. However, there is uncertainty as to what the information really means. The degree of uncertainty or confidence associated with a piece of information acts as a weight determining the degree of persuasion of the message. There are three main determinants of the uncertainty associated with a message: the clarity of the message, the consistency with other information, and the strength of the source commitment.

Clarity of the message. Personal influence will be more readily accepted if the signals sent by the source are high in clarity. High in clarity refers here to the lack of ambiguity in the communication, or the absence of noise in the communication (Shannon and Weaver 1949). As indicated earlier, in terms of the power of the source, technical expertise is not sufficient for information to be used by the recipient. Adapting the message so that the information is communicated at the technical level at which the recipient is knowledgeable and comfortable leads to a clearer message and therefore to a lower uncertainty associated with the message.

Consistency with other information. The degree to which information from one source is compatible with other information that the receiver has is a determinant of influence potential. The consistency of the information with other information enhances the acceptance of the information by the receiver (Howell and Burnett 1978). In fact, the recipient will evaluate new information against the prior opinion and the strength of (uncertainty associated with) this opinion (Wyer 1974). The consistency is important since previous information creates expectations and individuals react differently to confirmation or disconfirmation of expectations (Oliver 1977).

Strength of source commitment. The uncertainty attached to the message depends on the strength with which the information giver has committed himself or herself to the recommended behavior. The greater the commitment that the influencer has made to the recommended behavior, the greater the acceptance of the recommendation by the receiver. For example, if the source has bought the technology, the weight given to the information by the receiver will be greater. This, however, depends also on the type of information being given.

Positive information might be discounted, particularly, if the source has committed to the innovation. The reason for the discounting lies in the possibility of attributing the information to the source's motivation to reduce cognitive dissonance. This explains why negative information is more important in influencing the decision to adopt an innovation. In addition, the role of negative opinions of experts in blocking diffusion has been empirically demonstrated (Leonard-Barton 1985), as well as the importance of negative information when making a decision to adopt under time pressure or distraction (Wright 1974). Therefore, the nature of the information and whether it supports or negates an adoption decision has a major impact on the degree of influence.


This paper has developed a basic concept of diffusion theory: interpersonal influence. The proposed model offers explanations as to (1) why individuals give information to others, and (2) why individuals take information from others. The model also predicts the extent to which the information communicated influences recipient behavior.

We have systematically integrated a set of factors into a model explaining and predicting interpersonal communication effectiveness. This element of diffusion theory is crucial to understanding the rate of adoption of innovations, when affected by social imitation.

This model is the first stage in a research program centered on the role of interpersonal influence in the diffusion of innovations. The model remains to be tested. This can proceed in two complementary directions. The first direction consists of separately testing each of the relationships hypothesized in this paper. The second direction is an overall test of the model. Although the first approach would involve experimental research designs in the laboratory, the overall test of the model should be performed in the field to take advantage of the natural variation in the constructs of the model. This would also enhance the external validity of the model.


Aaker, David A. and John G. Myers (1982), Advertising Management, Englewood Cliffs, N.J.: Prentice-Hall, Inc.

Barone, Robert A. and Donn Byrne (1984), Social Psychology: Understanding Human Interaction, Boston: Allyn and Bacon, Inc.

Bass, Frank M. (1969), A New Product Growth Model for Consumer Durables," Management Science, 15 (S), 215-227.

Blau, Peter M. (1964), Exchange and Power in Social Life, New York: Wiles.

Blau, Peter M. (1974), On the Nature of Organizations, New York: Wiley.

Brock, T. (1964), Communication-Recipient Similarity and Decision Change," Journal of Personality and Social Psychology, 1, 650-54.

Czepiel, John A. (1975), "Word-of-Mouth Processes in the Diffusion of a Major Technological Innovation," Journal of Marketing Research, 11 (May), 172-180.

Gatignon, Hubert and Thomas S. Robertson (1985), "A Propositional Inventory for new Diffusion Research," Journal of Consumer Research, 11 (March), 849-367.

Gatignon, Hubert (1984), "Toward a Methodology for Measuring Advertising Copy Effects," Marketing Science, 3, 4 (Fall), 308-326.

Granovetter, Mark S. (1973), The Strength of Weak Ties," American Journal of Sociology, 78, 6, 1361-1380.

Granovetter, Mark S. (1983), The Strength of Weak Ties: A Network Theory Revisited, Sociological Theory, 1, 201-233.

Howell, William C., and Sarah A. Burnett (1978), "Uncertainty Measurement: A Cognitive Taxonomy, Organizational Behavior and Human Performance, 22, 45-68.

Jacoby, Jacob and Wayne D. Hoyer (1980), What If Opinion Leaders Didn't Really Know More" A Question of Nomological Validity, in Advances in Consumer Research, Vol. 8, ed. Kent B. Monroe, Association for Consumer Research, 299-302.

Katz, Elihu and Paul F. Lazarsfeld (1955), Personal Influence, New York: The Free Press.

Krugman, Herbert E. (1965), The Impact of Television Advertising: Learning Without Involvement," Public Opinion Quarterly, 29, 349-356.

Leonard-Barton, Dorothy (1985), Experts as Negative Opinion Leaders in the Diffusion of a Technological Innovation, Journal of Consumer Research, 11 (March), 914-926.

Oliver, Richard L. (1977), "Effect of Expectation and Disconfirmation on Past Exposure Product Evaluations: An Alternative Interpretation, Journal of Applied Psychology, 62, 4, 480-486.

Ray, Michael L. (1973), Marketing Communication and the Hierarchy of Effects, in new Models for Mass Communication Research, Vol. 2, 3d Peter Clarke, Beverley Hills: Sage, 147-476.

Robertson, Thomas S. (1976), Low-Commitment Consumer Behavior," Journal of Advertising Research, 16 (April), 19-24.

Rogers, Everett M., Hugh M. Daley and Thomas D. Wu (1982), The Diffusion of Home Computers: An Exploratory Study, Stanford: Institute for Communication Research, Stanford University.

Shannon, Claude E. and Warren Weaver (1949), The Mathematical Theory of Communication, Urbana, IL: University of Illinois Press.

Sternthal, Brian and C. Samuel Craig (1982), Consumer Behavior: An Information Process Perspective, Englewood Cliffs, NJ: Prentice-Hall.

Sternthal, Brian, Ruby Dholakia and Clark Leavitt (1978), The Persuasive Effect of Source Credibility: Tests of Cognitive Response, Journal of Consumer Research, 4, (March). 252-260.

Weimann, Gabriel (1982), On the Importance of Marginality: One More Step in the Two-Step Flow of Communication, American Sociological Review, 47 (December), 764-773.

Wright, Peter (1974), "The Harassed Decision Maker: Time Pressures, Distractions, and the Use of Evidence," Journal of Applied Psychology, 59, 5, 555-561.

Wyer, Robert S. (1974), Cognitive Organization and Change: An Information Processing Approach, Hillsdale, NJ: Lawrence Erlbaum Associates, Inc.



Hubert Gatignon, The Wharton School, University of Pennsylvania
Thomas S. Robertson, The Wharton School, University of Pennsylvania


NA - Advances in Consumer Research Volume 13 | 1986

Share Proceeding

Featured papers

See More


Q8. Avatars, Consumers and Possession in Online Gaming

Feihong Hu, Lancaster University, UK
Xin Zhao, Lancaster University, UK
Chihling Liu, Lancaster University, UK

Read More


System Justification and the Preference for Atavistic Products

Minju Han, Yale University, USA
George Newman, Yale University, USA

Read More


Doing Worse by Doing Good: How Corporate Social Responsibility makes Products Less Dangerous

Linda Lemarié, University of Neuchâtel
Florent Girardin, University of Neuchâtel

Read More

Engage with Us

Becoming an Association for Consumer Research member is simple. Membership in ACR is relatively inexpensive, but brings significant benefits to its members.